Welcome back to SavvyMonk, your one-stop for AI and tech news that actually matters.

Two years ago, Sam Altman told a podcast audience that a single person using AI would eventually build a billion-dollar company. A lot of people nodded along and filed it under "cool idea, maybe someday." Matthew Gallagher didn't wait for someday.

Let's get into it.

Ready to transform artificial intelligence from a buzzword into your personal revenue generator?

HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.

Inside you'll discover:

Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.

Get Your Guide

Matthew Gallagher is 41, self-taught, and until recently wasn't exactly a household name in tech.

His previous company, Watch Gang, sold wristwatches subscriptions and had 60 employees. It never turned a profit. More people meant higher costs and slower decisions, and Gallagher walked away from that experience with a clear lesson, headcount is a liability until it isn't.

In September 2024, he sat down with $20,000 and a stack of AI tools and decided to find out how lean a company could actually get.

Two months later, Medvi was live. It's a telehealth platform selling GLP-1 weight-loss drugs online, competing directly with Hims, Ro, and the rest of the compounded medication telehealth space. He didn't build a drug company. He built an interface, plugged it into existing infrastructure, and ran everything else with AI.

How He Built It

The AI stack did the work that would normally require an entire team. Gallagher used ChatGPT, Claude, and Grok to write the code powering Medvi's platform, generate website copy, produce images and videos for ads, and handle customer service. He built AI systems to analyze the business's performance in real time.

For visual content, Midjourney and Runway handled image and video generation. He even created an AI clone of his own voice to manage scheduling calls, freeing up more time to work.

For everything he couldn't do himself, he outsourced strategically rather than hiring.

CareValidate and OpenLoop Health handle the actual medical infrastructure: the doctors, pharmacies, shipping, and compliance. Gallagher's focus stayed entirely on branding, marketing, and customer acquisition. That's the architecture in one sentence: plug into existing clinical rails, use AI to run everything above them, and keep the team as small as possible.

His only full-time hire is his younger brother Elliot. He brought on a few contract engineers after an incident in March 2025 when a website change he made while out hiking broke production for an hour, costing around 200 customers. Rather than hire a full on-call team, he contracted two engineers. Still no headcount in the traditional sense.

The Numbers

Medvi launched with 300 customers in its first month. By month two, that number grew by another 1,000.

The New York Times was given direct access to Medvi's financials to verify the figures, and what they found was hard to ignore, $401 million in sales in 2025, its first full year in business, with 250,000 customers served and a net profit margin of 16.2%, which works out to roughly $65 million in profit.

For comparison, Hims and Hers posted a 5.5% net margin the same year with over 2,400 employees. Medvi's current projection is $1.8 billion in 2026 revenue, with two employees.

Gallagher described the growth as "insane." Altman emailed after the story ran to say he appeared to have won a bet with his tech CEO friends over when exactly a company like this would appear, and that he'd like to meet the guy.

The Complications You Should Know About

The NYT profile, published April 2, was widely shared as a clean success story. But six weeks before that profile ran, the FDA had already sent Medvi a warning letter.

On February 20, 2026, the agency flagged misbranding violations on Medvi's website, specifically that the site language falsely implied FDA approval of the compounded GLP-1 products it sold, and suggested Medvi was the compounder of those products when it isn't. The FDA warned that failure to fix the violations could lead to seizure or injunction.

It's worth putting this in context, Medvi wasn't singled out.

In September 2025, the FDA sent warning letters to more than 50 GLP-1 compounders and telehealth companies for similar marketing violations, and issued another batch to more than 30 companies in March 2026. This is an industry-wide enforcement pattern, not a targeted investigation. Medvi is operating in a sector that regulators are actively scrutinizing, and the NYT's decision not to mention the warning letter is an editorial choice readers should be aware of.

There's also the OpenLoop Health data breach to note. OpenLoop, one of Medvi's two clinical infrastructure partners, disclosed in January 2026 that a threat actor had breached its systems and claimed to have exfiltrated records from approximately 1.6 million patients, including names, contact information, dates of birth, and medical information. Medvi's patients are among those in the OpenLoop network.

None of this makes the Medvi story less impressive. It does make it more complete.

Why This Actually Matters

Critics have noted that Gallagher's success owes as much to market timing as to the AI toolkit. He entered a sector with massive unmet demand, GLP-1 drugs that people desperately want and that traditional healthcare makes hard to access, and used AI to move faster than anyone else. That's a fair observation. The market did a lot of the heavy lifting on demand.

But the operational model is genuinely new. One person using AI to run code, copy, ads, customer service, and analytics simultaneously, while outsourcing compliance and fulfillment to specialized partners, is not how companies used to get built. The traditional startup playbook requires a team for each of those functions. Gallagher compressed all of them into a toolset and a very long workday.

OpenLoop's CEO Jon Lensing, one of his key infrastructure partners, put it this way when talking to the Times: Gallagher's "native tongue seems to be A.I." That framing matters. This isn't a story about someone who stumbled onto a hot drug market. It's about someone with genuine technical fluency who understood exactly how to use AI as a force multiplier before most founders were thinking that way.

Gallagher is now expanding. Medvi's men's health line launched in February and reached 50,000 customers in its first month. Women's health and hormone therapy are next, along with healthy meal delivery. The company now reports over 500,000 patients. He's also donated $1 million to the Gallagher Foundation, which he set up to fund entrepreneurs using their skills for nonprofits and public benefit.

The Bottom Line

Sam Altman's 2024 prediction turned out to have a 9-month runway, not a three-year one.

Medvi is the first clean proof point that AI can genuinely replace most of a startup's operational workforce, not in theory, but in practice, with verified financials and a profit margin that embarrasses its much larger competitors.

The regulatory picture adds real risk, and that risk is worth watching closely. But the blueprint itself is already being replicated by founders who read this story and started counting their own AI tokens.

Category: Business Strategy

"I'm building a lean company in [industry] with minimal headcount. My core value proposition is [describe it in one sentence]. Help me map out which business functions I could automate with AI tools versus which ones I need to outsource to a specialized partner, and which ones genuinely require a human hire. For each function, suggest a specific tool, platform, or type of vendor I could use."

Gallagher mentioned he's considering hiring more people mostly because he's lonely. There's something genuinely worth sitting with there. Maximum efficiency and isolation often show up together. The thing that makes this company possible, a single person with AI running everything, is also the thing that makes it strange to live inside. The business case for staying small is airtight. The human case is still being figured out.

Hit reply, I read every response.

See you in the next one.

— Vivek

P.S. If you know someone who thinks about entrepreneurship, startups, or what AI is actually changing in the real world, this one's worth forwarding. They can subscribe at https://savvymonk.beehiiv.com/